Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Games

Benchmarks

Task NameDataset NameSOTA ResultTrend
RecommendationGames
Precision@202.13
30
Top-K RecommendationGames
MRR@205.32
30
Next-Item RecommendationGames Amazon (test)
HR@100.782
27
RecommendationGames
HR@163.26
19
Sequential RecommendationGames
NDCG@100.0757
17
Sequential RecommendationGames (test)
NDCG@55.6
15
Generative RecommendationGames
Recall @ 50.0612
15
RecommendationGames
Recall@102.13
12
Sequential RecommendationGames
HR@137.9
12
Sequential RecommendationGames Noisy (20% noise) (test)
HR@103.14
12
Sequential RecommendationGames Clean (test)
HR@105.34
12
Session RecommendationGames
HR@50.6
11
Sequential RecommendationGames
HR@100.1041
11
Next-item predictionGames 100 negative samples
HR@1081.9
10
Next-item predictionGames (test)
HR@1014.1
9
Sequential RecommendationGames
Average Batch Runtime (s)0.008
9
Generative RecommendationGames
Recall@50.0338
8
RecommendationGames
Recall@1015.16
7
RecommendationGames strong generalization
Recall@2029.98
7
DiagnosticHeld-out games (test)
Quality Score6.07
6
RecommendationGames
nDCG@1023.164
6
GameCWM Generationgames (held-out)
Mean Verification Score66.7
5
RecommendationGames
HR@53.12
5
Schema MatchingGames
F1 Score100
4
End-to-End Data IntegrationGames
Output Records65,794
2
Showing 25 of 26 rows